J. J. Gray

Reliable protein-protein docking with AlphaFold, Rosetta and replica-exchange

This paper improves protein complex structure prediction by combining AlphaFold with replica-exchange docking, achieving better results on challenging cases.

Implicit model to capture electrostatic features of membrane environment.

This paper introduces an implicit electrostatic model for membrane proteins that improves tilt angle calculations and enables faster design calculations.

Structure-Based Neural Network Protein-Carbohydrate Interaction Predictions at the Residue Level

The paper introduces two deep learning models to predict carbohydrate binding sites on proteins, with one model outperforming the other and showing effectiveness on both experimental and predicted structures.

Virtually the same? Evaluating the effectiveness of remote undergraduate research experiences

This study evaluates remote undergraduate research experiences during the pandemic, finding similar scientific integration outcomes to in-person programs but with differences based on prior experience.

Generative language modeling for antibody design

This paper introduces IgLM, a generative language model for designing synthetic antibody sequences with improved developability.

Improved antibody structure prediction by deep learning of side chain conformations

This paper introduces DeepSCAb, a deep learning method that improves antibody structure prediction by modeling side chain conformations, outperforming existing methods that lack side chain context.

Affinity-engineered human antibodies detect celiac disease gluten pMHC complexes and inhibit T-cell activation

The paper describes the engineering of high-affinity human antibodies that recognize gluten epitopes bound to HLA-DQ2.5 and inhibit T-cell activation in celiac disease.

Accurately positioning functional residues with robotics-inspired computational protein design

This paper introduces two computational methods for accurately positioning functional residues in proteins and validates them by redesigning enzyme active sites.

Antibody structure prediction using interpretable deep learning

This paper introduces DeepAb, an interpretable deep learning method for predicting antibody FV structures from sequence, showing improved accuracy and providing insights into residue interactions.

"How do we do this at a distance?!" A descriptive study of remote undergraduate research programs during COVID-19

This study describes and evaluates remote undergraduate research programs in life sciences during the COVID-19 pandemic, identifying strengths, weaknesses, and recommendations for improvement.

Colicin-mediated transport of DNA through the iron transporter FepA

This study shows colicin B uses the iron transporter FepA to transport DNA into E. coli, revealing a mechanism for cargo delivery through bacterial membranes.

Structural basis for peptide substrate specificities of glycosyltransferase GalNAc-T2

This study identifies sequence and structural motifs that determine peptide substrate specificity for GalNAc-T2, revealing key residues and conformations that influence glycosylation efficiency.

Diverse scientific benchmarks for implicit membrane energy functions

The paper introduces 12 benchmarks to evaluate implicit membrane energy functions using diverse datasets and demonstrates their application with the franklin2019 energy function.

Robustification of RosettaAntibody and Rosetta SnugDock

This paper improves RosettaAntibody and Rosetta SnugDock by simplifying interfaces, expanding templates, and enhancing docking accuracy through benchmarking.

Geometric Potentials from Deep Learning Improve Prediction of CDR H3 Loop Structures

DeepH3 improves CDR H3 loop structure prediction by learning geometric potentials from antibody sequences, outperforming standard Rosetta energy functions in benchmark tests.

Protein Docking and Steered Molecular Dynamics Reveal Alternative Regulatory Sites on the SERCA Calcium Transporter

This study uses SMD and docking to test if PLB interacts with alternative sites on SERCA beyond the canonical site, finding that the canonical site is preferred but alternative sites are favorable under certain conditions.

Novel sampling strategies and a coarse-grained score function for docking homomers, flexible heteromers, and oligosaccharides using Rosetta in CAPRI Rounds 37-45

The paper improves Rosetta docking methods for complex targets in CAPRI Rounds 37-45, achieving successful predictions and refining models for symmetric and carbohydrate-containing complexes.

Toward computational design of protein crystals with improved resolution

This paper develops a computational method to improve protein crystal resolution by designing point mutations, finding that resolution can be enhanced and space group altered through such mutations.

Protein structure prediction and design in a biologically-realistic implicit membrane

This paper develops an implicit membrane model for protein structure prediction and design, improving accuracy in membrane protein properties and sequence recovery.

Flexible backbone assembly and refinement of symmetrical homomeric complexes

The paper introduces Rosetta SymDock2, a method for modeling symmetric protein complexes from homology-modeled monomers, achieving better accuracy than existing algorithms.

Repertoire analysis of antibody CDR-H3 loops suggests affinity maturation does not typically result in rigidification

The paper analyzes the flexibility of CDR-H3 loops in antibodies to assess if affinity maturation leads to rigidification, finding only a slight decrease in flexibility.

Efficient Flexible Backbone Protein-Protein Docking for Challenging Targets

This paper introduces a new protein-protein docking method that efficiently samples flexible backbone conformations and achieves higher success rates for docking flexible complexes.

Comparison Of NMR And Crystal Structures Of Membrane Proteins And Computational Refinement To Improve Model Quality

The paper compares NMR and crystal structures of membrane proteins, identifies structural differences, and proposes refinement methods to improve NMR model quality.

The Rosetta all-atom energy function for macromolecular modeling and design

The paper explains the Rosetta all-atom energy function, its mathematical models, and its application to biomolecular modeling and design, including new capabilities for membrane proteins and other macromolecules.